Auditory evoked potentials from deaf individuals using cochlear implants

Abstract

Auditory evoked potentials (AEPs) provide an objective measure central electrode where the grand average amplitudes were largest period (HIGH: mean 244.78±50.18 months, LOW: mean of auditory cortical function. However AEPs from cochlear i l ( ) i d b h l i l f for both groups (approximately FCz). AEP peak amplitudes and l i d i d i i i d 532.44±34.16 months; t(16)=-4.74, p<.001) implant CI) users are contaminated by the electrical artefact d d b th d i I d d t t l i (ICA) latencies were determined using a semi-automatic procedure as produced by the device. Independent component analysis i l t d i kd t ( bill il/ kd t ht l) h b t d t tt t th CI t f t d th implemented in peakdet.m www.billauer.co.peakdet.html). has been reported to attenuate the artefact and recover the AEPs. Here the effects of CI artefact attenuation on the quality of Results the AEPs were systematically investigated Electroencephalogram investigated. data were recorded from 18 adult CI users presented with ICA sensitivity- there was not a single CI user dataset where AEPs auditory and visual stimuli. The CI artefact attenuation rate was were not buried by large electrical CI artefact. calculated to investigate ICA sensitivity and AEP quality was Figure 1 – Cochlear implant (CI) artefact evoked by determined based on a signal to noise ratio ( SNR) measure. ICA g ) different environmental sounds. Panels in the top row specificity was evaluated with a hybrid simulation approach and show the sound profile of two stimuli used in the p y y pp experiment, on the left side a “sheep” sound, on the by comparing visual evoked potentials (VEPs) from CI users with sheep right side a “hitting baseball bat” sound. The panels in and without CI artefact attenuation. The results showed that the middle row show the root mean square (RMS) of the EEG activity of two CI users when stimulated with AEPs could be recovered from all CI users, indicating high i i i li d hi hl l d y f the two sounds, respectively. Both implantees used a sensitivity. Moreover, AEP amplitudes were highly correlated N l CI t th i ht id ith th ESP it G i h d i h i di id l diff ll Nucleus system on the right side with the ESPrit processor and the ACE speech processing strategy. The Figure 5 – Comparison of clinical profiles from cochlear implant (CI) users with low (light grey) and high (dark grey) with age, demonstrating that individual differences were well d CI ith hi h AEP SNR h t i d b p p p g gy topography at the latency of maximum amplitude is illustrated for the same subject on both panels The SNR AEPs. All panels show mean ± SEM values. A) Self-reported duration of deafness in months. B) Age at preserved. users with a high were characterised by a implantation in months i e age when the CI device was switched on C) Experience with the CI device in months D) significantly shorter duration of deafness compared to low AEP panels. panels in the bottom row show RMS activity for months, i.e. on. months. Speech scores: percentage correct on the Bamford-Kowal-Bench (BKB) speech recognition test in quiet. SNR individuals The results confirm that ICA is a valid tool to another two CI users and the respective topography at individuals. the latency of maximum amplitude for one of them attenuate CI artefacts allowing the objective non-invasive study them. Both implantees used the Advanced Bionics CI system ICA specificity: CI users VEPs evaluation artefacts, objective, d l of auditory cortex rehabilitation in CI users on the left side with the AB Harmony processor and the HiRes-S with Fidelity 120 speech processing strategy CI-corrected VEPs were minimal users.

More information

Identifiers

ORCID for S. Bleeck: orcid.org/0000-0003-4378-3394

Catalogue record

Export record